By using empirical methods as employed in pure sciences where repeatability of results is a test for the veracity of any formula or theory, the Gettier problem can be resolved. The resolution however should begin with first arriving at a consensus on what it means for an individual to believe in something and be justified in doing so. One can not form a belief based on hasty generalization. It was just a coincidence that the time was actually true but coincidences do not make the rule or knowledge. The characters in Gettier examples do not have knowledge but a semblance or verisimilitude of knowledge on the basis of coincidental truths.
A solution to the Gettier’s counterexample in this case would be that if I were to look at the clock some time later and having found the clock displaying the same time, I would have known my earlier knowledge as being based on a coincidental truth and thus only a false appearance of knowledge. While the time displayed was coincidentally true, I can not form a belief that this is “knowledge” or I “know” the right time without considering the element of probability that the clock may not be running or displaying wrong time. The Gettier counter-examples to justified true belief are thus valid and applicable in all cases where elements of coincidence or probability are involved. A repeatability of observations in most cases as mentioned in the solution to Gettier counter examples would reveal the element of chance that may have at first been overlooked.
Besides the three criteria of justified true belief, for something to be classified as knowledge one needs to consider all the factors that may be at play. May be the human mind is not sophisticated enough to grasp all of them or may be it gets vainglorious enough to think that it has considered all the factors. A framework that accommodates all these factors must incorporate a fourth entity based on probability theory that takes care of these coincidental truths. “To know something” and “to know something for sure” requires eliminating the elements of probability that keep throwing up some inconsistencies in an otherwise working definition of justified true belief as knowledge.
Lemos, N.M. “The traditional analysis and gettier problem.“ An introduction to the theory of knowledge. Cambridge University Press, 2007. 22-43.